[Serve][LLM] Fix malformed PromQL in Serve LLM Grafana dashboard

[eicherseiji]

Why are these changes needed?

The Serve LLM Grafana dashboard (generate_serve_llm_grafana_dashboard) emits PromQL label matchers with a stray comma when the optional {global_filters} substitution is empty (the default). With no global filters set, the generated queries contain:

• 32 leading-comma matchers, e.g. ray_serve_deployment_request_counter_total{, deployment=~"$deployment"}
• 2 double-comma matchers, e.g. ... WorkerId=~"$workerid", , deployment=~"$deployment"

Prometheus rejects both:

bad_data: invalid parameter "query": parse error: unexpected "," in label matching, expected identifier or "}"

Root cause: _WORKER_JOIN and _VLLM_DEPLOYMENT_FILTER placed {global_filters} in a non-trailing position, so an empty substitution left a leading or double comma. The other fragments (_VLLM_FILTER, _WORKERID_FILTER) already keep {global_filters} trailing, where an empty substitution yields only a trailing comma (which Prometheus tolerates).

This moves {global_filters} to the trailing position in the two offending fragments to match that convention.

Fixes #63219

Related issue number

Closes #63219

Checks

• I've signed off every commit (-s) per the DCO requirement.
• ruff/black pass via pre-commit on the changed file.

[gemini-code-assist]

Code Review

This pull request repositions the {global_filters} placeholder to the end of the Prometheus metric filter lists in serve_llm_dashboard_panels.py to prevent syntax errors (like leading or double commas) when the substitution is empty. There are no review comments, and I have no feedback to provide.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.